P. S. Hiremath et al.

, International Journal of Advances in Computer Science and Technology, 2(12), December 2013, 267-274

267



ABSTRACT

In spite of over two decades of intense research,
illumination and pose invariance remain prohibitively
challenging aspects of face recognition for most practical
application. Many recent events, such as terrorist attacks,
have exposed the serious weaknesses in most sophisticated
security systems. Automatic face recognition has long been
established as one of the most active research areas in
computer vision. In spite of the large number of developed
algorithms, real-world performance of state-of-the-art
methods has been disappointing. Three dimensional (3D)
human face recognition is emerging as a significant biometric
technology. Research interest in 3D face recognition has
increased during recent years due to the availability of
improved 3D acquisition devices and processing algorithms.
In this paper, we have proposed a novel method for three
dimensional (3D) face recognition using Radon transform
and Symbolic LDA based features of 3D face images. In this
method, the Symbolic LDA based feature computation takes
into account the face image variations to a larger extent and
has the advantage of dimensionality reduction. The
experimental results have yielded 99.60% recognition
accuracy using SVM with reduced computational cost, which
compares well with other state-of-the-art methods.

Key words : 3D face recognition, radon transform, symbolic
LDA, KNN, SVM.

1. INTRODUCTION

Biometrics refers to the identification of humans by their
characteristics or traits. Biometrics is used in computer
science as a form of identification and access control. It is also
used to identify individuals in groups that are under
surveillance. Many different aspects of human physiology,
chemistry or behaviour can be used for biometric
authentication. The selection of a particular biometric for use
in a specific application involves a weighting of several
factors. A number of biometric traits have been developed and
are used to authenticate the person's identity. The idea is to
use the special characteristics of a person to identify him. By
using special characteristics we mean the using the features
such as face, iris, fingerprint, signature etc.


The method of identification based on biometric
characteristics is preferred over traditional passwords and
PIN based methods for various reasons such as: The person to
be identified is required to be physically present at the
time-of-identification. Identification based on biometric
techniques obviates the need to remember a password or carry
a token. A biometric system is essentially a pattern
recognition system which makes a personal identification by
determining the authenticity of a specific physiological or
behavioural characteristic possessed by the user. A biometric
system can be either an 'identification' system or a
'verification' (authentication) system.

Human face images are useful not only for person
recognition, but for also revealing other attributes like gender,
age, ethnicity, and emotional state of a person. Therefore, face
is an important biometric identifier in the law enforcement
and human computer interaction (HCI) communities.
Detecting faces in a given image and recognizing persons
based on their face images are classical object recognition
problems that have received extensive attention in the
computer vision literature. While humans are perceived to be
good at recognizing familiar faces, the exact cognitive
processes involved in this activity are not well understood.
Therefore, training a machine to recognize faces as humans
do is an arduous task. However, general methods used in
object recognition such as appearance based, model based,
and texture based approaches are also applicable to the
specific problem of face detection and recognition.

The face is the frontal portion of the human head,
extending from the forehead to the chin and includes the
mouth, nose, cheeks, and eyes. Being the foremost part in
ones interactions with the outer world, the face houses most
of the fundamental sensory organs necessary for perceiving
the world around, namely, eyes for seeing, nose for smelling,
mouth for tasting, and ears for hearing. The face is considered
to be the most commonly used biometric trait by humans; we
recognize each other and, in many cases, establish our
identities based on faces. Hence, it has become a standard
practice to incorporate face photographs in various tokens of
authentication such as ID cards, passports, and drivers
licenses.
Face Recognition and verification have been at the top of
the research agenda of the computer vision community for
more than a decade. The scientific interest in this research
topic has been motivated by several factors. The main
RADON TRANSFORM AND SYMBOLIC LINEAR DISCRIMINANT
ANALYSIS BASED 3D FACE RECOGNITION USING KNN AND SVM
P. S. Hiremath, Manjunatha Hiremath
1

1
Department of Computer Science,
Gulbarga University, Gulbarga, Karnataka, India
hiremathps53@yahoo.com, manju.gmtl@gmail.com

ISSN 2320 - 2602
Volume 2, No.12, December 2013
International Journal of Advances in Computer Science and Technology
Available Online at http://warse.org/pdfs/2013/ijacst022122013.pdf

P. S. Hiremath et al., International Journal of Advances in Computer Science and Technology, 2(12), December 2013, 267-274

268


attractor is the inherent challenge that the problem of face
image processing, face detection and recognition. However,
the impetus for better understanding of the issues raised by
automatic face recognition is also fuelled by the immense
commercial significance that robust and reliable face
recognition technology would entail. Its applications are
envisaged in physical and logical access control, security,
man-machine interfaces and low bitrate communication.

To date, most of the research efforts, as well as commercial
developments, have focused on two dimensional (2D)
approaches. This focus on monocular imaging has partly been
motivated by costs but to a certain extent also by the need to
retrieve faces from existing 2D image and video database.
With recent advances in image capture techniques and
devices, various types of face-image data have been utilized
and various algorithms have been developed for each type of
image data. Among various types of face images, a 2D
intensity image has been the most popular and common
image data used for face recognition because it is easy to
acquire and utilize. It, however, has the intrinsic problem that
it is vulnerable to the change of illumination. Sometimes the
change of illumination gives more difference than the change
of people, which severely degrades the recognition
performance. Therefore, illumination-controlled images are
required to avoid such an undesirable situation when 2D
intensity images are used. To overcome the limitation of 2D
intensity images, Three Dimensional (3D) images are being
used, such as 3D meshes and range images. A 3D mesh image
is the best 2D representation of 3D objects. It contains 3D
structural information of the surface as well as the intensity
information of each point. By utilizing the 3D structural
information, the problem of vulnerability to the change of
illumination can be solved. A 3D mesh image is suitable
image data for face recognition, but it is complex and difficult
to handle.

A range image is simply an image with depth information.
In other words, a range image is an array of numbers where
the numbers quantify the distances from the focal plane of the
sensor to the surfaces of objects within the field of view along
rays emanating from a regularly spaced grid. Range images
have some advantages over 2D intensity images and 3D mesh
images. First, range images are robust to the change of
illumination and colour because the value on each point
represents the depth value which does not depend on
illumination or colour. Also, range images are simple
representations of 3D information. The 3D information in 3D
mesh images is useful in face recognition, but it is difficult to
handle. Different from 3D mesh images, it is easy to utilize
the 3D information of range images because the 3D
information of each point is explicit on a regularly spaced grid.
Due to these advantages, range images are very promising in
face recognition.

The majority of the 3D face recognition studies have
focused on developing holistic statistical techniques based on
the appearance of face range images or on techniques that
employ 3D surface matching. A survey of literature on the
research work focusing on various potential problems and
challenges in the 3D face recognition can be found in the
survey[1-5,16,17]. Gupta et al.[6] presented a novel
anthropometric 3D face recognition algorithm. This approach
employs 3D Euclidean and Geodesic distances between 10
automatically located anthropometric facial fiducial points
and a linear discriminant classifier with 96.8% recognition
rate. Lu et al.[7] constructed many 3D models as registered
templates, then they matched 2.5D images (original 3D data)
to these models using iterative closest point (ICP). Chang et al.
[8] describe a multi-region approach to 3D face recognition.
It is a type of classifier ensemble approach in which multiple
overlapping sub regions around the nose are independently
matched using ICP and the results of the 3D matching are
fused. Jahanbim et al. [9] presented an approach of
verification system based on Gabor features extracted from
range images. In this approach, multiple landmarks (fiducials)
on face are automatically detected, and also the Gabor
features on all fiducials are concatenated, to form a feature
vector to collect all the face features. Hiremath et al.[18] have
discussed the 3D face recognition by using Radon Transform
and PCA with recognition accuracy of 95.30%. Hengliand
Tang et al.[19] presented a 3D face recognition algorithm
based on sparse representation. In this method they used
geo-metrical features, namely, triangle area, triangle normal
and geodesic distance.

In this paper, our objective is to propose Radon transform
and Symbolic LDA based 3D face recognition. The
experimentation is carried out using three publicly available
databases, namely, Bhosphorus, Texas and CASIA 3D face
databases.

2. MATERIALS AND METHODS

For purpose of experimentation of the proposed
methodology, the face images drawn from the following 3D
face databases are considered: (i) Boshphorus 3D face
database, (ii) Texas 3D face database, (iii) CASIA 3D face
database.

2.1 Bosphorus 3D Face Database

The Bosphorus 3D face database consists of 105 subjects in
various poses, expressions and occlusion conditions. The 18
subjects have beard/moustache and the 15 subjects have hair.
The majority of the subjects are aged between 25 and 35.
There are 60 men and 45 women in total, and most of the
subjects are Caucasian. Two types of expressions have been
considered in the Bosphorus database. In the first set, the
expressions are based on action units. In the second set, facial
expressions corresponding to certain emotional expressions
are collected. These are: happiness, surprise, fear, sadness,
anger and disgust.

P. S. Hiremath et al., International Journal of Advances in Computer Science and Technology, 2(12), December 2013, 267-274

269


The facial data are acquired using Inspeck Mega Capturor
II 3D, which is a commercial structured-light based 3D
digitizer device. The sensor resolution in x, y & z (depth)
dimensions are 0.3mm, 0.3mm and 0.4mm respectively, and
colour texture images are high resolution (1600x1200 pixels).
It is able to capture a face in less than a second. Subjects were
made to sit at a distance of about 1.5 meters away from the 3D
digitizer. A 1000W halogen lamp was used in a dark room to
obtain homogeneous lighting. However, due to the strong
lighting of this lamp and the devices projector, usually
specular reflections occur on the face. This does not only
affect the texture image of the face but can also cause noise in
the 3D data. To prevent it, a special powder which does not
change the skin colour is applied to the subjects face.
Moreover, during acquisition, each subject wore a band to
keep his/her hair above the forehead to prevent hair occlusion,
and also to simplify the face segmentation task. The propriety
software of the scanner is used for acquisition and 3D model
reconstruction[10].

2.2 Texas 3D Face Database

The Texas 3D Face Recognition (Texas 3DFR) database is
a collection of 1149 pairs of facial color and range images of
105 adult human subjects. These images were acquired using
a stereo imaging system manufactured by 3Q Technologies
(Atlanta, GA) at a very high spatial resolution of 0.32 mm
along the x, y, and z dimensions. During each acquisition, the
color and range images were captured simultaneously and
thus the two are perfectly registered to each other. This large
database of two 2D and 3D facial models was acquired at the
company Advanced Digital Imaging Research (ADIR), LLC
(Friendswood, TX), formerly a subsidiary of Iris International,
Inc. (Chatsworth, CA), with assistance from research
students and faculty from the Laboratory for Image and Video
Engineering (LIVE) at The University of Texas at Austin.
This project was sponsored by the Advanced Technology
Program of the National Institute of Standards and
Technology (NIST). Texas 3DFRD was created to develop
and test 3D face recognition algorithms intended to operate in
environments with cooperative subjects, wherein, the faces
are imaged in a relatively fixed position and distance from the
camera [11].

2.3 CASIA 3D Face Database

CASIA 3D Face Database consisting of 4624 scans of 123
persons using the non-contact 3D digitizer, Minolta Vivid
910. During building the database, not only the single
variations of poses, but also expressions and illuminations are
considered [12].

3. PROPOSED METHOD

The proposed methodology employs the following:
(i) Radon transform(RT) and
(ii) Symbolic Linear Discriminant Analysis (Symbolic LDA),
which are described in the following sections.

3.1 Radon Transform

The radon transform (RT) is a fundamental tool in many
areas. The 3D radon transform is defined using 1D
projections of a 3D object f(x,y,z) where these projections are
obtained by integrating f(x,y,z) on a plane, whose orientation
can be described by a unit vector
r
. Geometrically, the
continous 3D radon transform maps a function
3
into the
set of its plane integrals in
3
. Given a 3D function
( ) ( , , ) f x f x y z
r
@ and a plane whose representation is
given using the normal
r
and the distance s of the plane
from the origin, the 3D continuous radon transform of f for
this plane is defined by
( , ) ( ) ( )
T
f a s f x x s dx


9 =
} } }
r r r r

( , , ) ( si n cos si n si n
cos )
f x y z x y
z s d x d y d z

+
=
+
} } }

where x
r
=[ , , ]
T
x y z ,

r
=[sin cos ,sin sin ,cos ]
T
, and is Diracs
delta function defined by ( ) 0, 0, x x = = ( ) 1 x dx

=
}
.
The radon transform maps the spatial domain (x,y,z) to the
domain ( ,s
r
) [13].

3.2 Symbolic Faces

Consider the 3D range face images
1 2
, ,...
n
I I I , each of
size N M , from a 3D face image database. Let
1 2
{ , ,... }
n
O = I I I be the collection of n face images of the
database, which are first order objects. Each object ,
l
I eO
1,2,..., l n = , is described by a feature vector
1
( ,..., )
p
Y Y
% %
,
of length p=NM, where each component
j
Y
%
, 1,2,..., j p = ,
is a single valued variable representing the range values of
the 3D face image
l
I . An image set is a collection of 3D
range face images of m different subjects; each subject has
same number of images but with different facial expressions
and illuminations. There are m number of second order
objects (face classes) denoted by
1 2
, ,...,
m
c c c ,each
consisting of different individual images
l
I eO . We
denote the set
1 2
{ , ,..., }
m
E c c c = and
i
c _ O ,
1,2,..., i m = . The feature vector of each face class
i
c E e
is described by a vector of p interval variables
1 2
, ,...,
p
Y Y Y ,
P. S. Hiremath et al., International Journal of Advances in Computer Science and Technology, 2(12), December 2013, 267-274

270


and is of length p NM = . The interval variable
j
Y of face
class
i
c is declared by ( ) [ , ]
j i ij ij
Y c x x = , where
ij
x and
ij
x are minimum and maximum intensity values,
respectively, among
th
j range values of all the images of face
class
i
c . This interval incorporates information of the
variability of
th
j feature inside the
th
i face class. We denote
( ) ( ( ),...,( ( ))
i l i p i
X c Y c Y c = . The vector ( )
i
X c of
symbolic variables is recorded for each
i
c E e , and can be
described by a symbolic data vector which is called as
symbolic face[14-17] :
1 2
( ) ( , ,..., )
i i i ip
X c a a a = , where
( )
j i
aij Y c = , 1,2,..., j p = . We represent the m symbolic
faces by a m p matrix :
11 1
1
...
. . . ( )
...
p
ij m p
m mp
a a
X a
a a

| |
|
= =
|
|
\ .

3.3 Symbolic LDA

We consider the extension of linear discriminant analysis
(LDA) to symbolic data analysis frame work [14-17].
Consider the 3D range face images
1 2
, ,...,
n
I I I , each of
size M x N, from Texas 3D range face image database. Let
1 2
, ,...,
n
O = I I I be the collection of n 3D range face
images of the database, which are first order objects. Each
object ,
l
I eO 1,2,..., l n = , is described by a matrix
l
A
( 1,2,..., ) l n = , where each component
, 1,2,..., ,
ab
Y a M =
%
and b=1,2,,N, is a single valued
variable representing the 3D range values of the face image
l
I . An image set is a collection of face images of m different
subjects and each subject has different images with varying
expressions and illuminations. Thus, there are m number of
second order objects (face classes) denoted by
1 2
{ , ,..., }
m
E c c c = , each consisting of different individual
images,
l
I eO, of a subject. The face images of each face
class are arranged from right side view to left side view. The
feature matrix of
th
k sub face class
k
i
c of
th
i face class
i
c ,
where 1,2,..., k q = , 1,2,..., i m = , is described by a
matrix
k
i
X of size M N that contains interval variable
k
iab
a , 1,2,..., a M = , and 1,2,..., b N = . The matrix is
called as symbolic face and is represented as :
11 1
1
...
. . .
....
k k
i i N
k
i
k k
iM iMN
a a
X
a a
(
(
=
(
(



The interval variable
k
iab
a of
th
k sub face class
k
i
c of
th
i face
class
i
c is described as ( ) [ , ]
k k k k
iab i iab iab
a c x x = , where
k
iab
x
and
k
iab
x are minimum and maximum intensity values,
respectively, among ( , )
th
a b feature inside the
th
k sub face
class of
th
i face class. Thus, we obtain the qm symbolic
faces from the given image database.
Now, we apply LDA method to the centers
k
iab
x e of the
interval [ , ]
k k
iab iab
x x given by
2
c
k k
k iab iab
iab
x x
x
+
=
The M N symbolic face
c
k
i
X containing the centers
c
k
iab
x R e of the intervals
k
iab
a of symbolic face
k
i
X is given
by
11 1
1
...
. . .
....
c c
c
c c
k k
i i N
k
i
k k
iM iMN
a a
X
a a
(
(
( =
(
(



In the symbolic LDA approach, to calculate the scatter
(within and between class) matrices of qm symbolic faces
k
i
X , where i=1,2,,m and k=1,2,,q, we define the
within-class image scatter matrix
w
S as
1 1
( ) ( )
c c
q m
k T k
w i i i i
i k
S X M X M
= =
=

where
1
1 c
q
k
i i
k
M X
q
=
=

, and the between-class image
scatter matrix
b
S as
1
( ) ( )
m
T
b i i
i
S M M M M
=
=

,
where
,
1 c
k
i
i k
M X
qm
=

. In discriminant analysis, we want
to determine the projection axis that maximizes the ratio
det{ }
det{ }
b
w
S
S
. In other words, we want to maximize the
between-class image scatter matrix while minimizing the
within-class image scatter matrix. It has been proved [10,14]
that this ratio is maximized when the column vector of
projection axis V is the eigenvector of
1
w b
S S

corresponding
to first p largest eigenvalues. For each symbolic face
k
i
X ,
the family of projected feature vectors,
1 2
, ,...,
p
Z Z Z are
considered as:
k
s i s
Z X V =
P. S. Hiremath et al., International Journal of Advances in Computer Science and Technology, 2(12), December 2013, 267-274

271


where s=1,2,,p. Let
1 2
[ , ,..., ]
k
i p
B Z Z Z = , which is
called as the feature matrix of the symbolic face
k
i
X . The
feature matrix
test
B of the test image
test
X is obtained as :
( ) test s test s
Z X V = ,
where s=1,2,,p and
( )1 ( )2 ( )
[ , ,..., ]
test test test test p
B Z Z Z = .

3.4 Proposed Method

The Figure 1 shows the overview of proposed framework.
The algorithms of the training phase and the testing phase of
the proposed method are given below:

Algorithm 1: Training Phase
1. Input the range image I
1
from the training set
containing M images.
2. Apply Radon transform, from 0
o
to 180
o

orientations (in steps of h), to the input range image
I
1
yielding a binary image I
2
.
3. Superpose the binary image I
2
obtained in the Step 2
on the input range image I
1
to obtain the cropped
facial range image I
3
.
4. Repeat the Steps 1 to 3 for all the M facial range
images in the training set.
5. Apply Symbolic LDA to the set of cropped facial
range.
6. Compute the weights
1
w ,
2
w , ...,
p
w for each
training face image, where p <M is the dimension of
eigen subspace on which the training face image is
projected.
7. Store the weights
1
w ,
2
w ,...,
p
w for each training
image as its facial features in the Symbolic LDA
feature library of the face database.

Algorithm 2: Testing Phase
1. Input the test range image Z
1
.
2. Apply Radon transform, from 0
o
to
180
o
orientations (in steps of h), to the input range
image Z
1
yielding a binary image Z
2
.
3. Superimpose the binary image Z
2
on Z
1
to obtain the
cropped facial image Z
3
.
4. Compute the symbolic weights , 1,2,...,
test
i
w i p = ,
for the test image Z
1
by projecting the test image on
the Symbolic LDA feature subspace of dimension p.
5. Compute the Euclidian distance D between the feature
vector
test
i
w and the feature vectors stored in the
Symbolic LDA feature library.
6. The face image in the face database, corresponding to
the minimum distance D computed in the Step 5, is
the recognized face.
7. Output the texture face image corresponding to the
recognized facial range image of the Step 6.
4. RESULTS AND DISCUSSION

The proposed method is implemented using Intel Core 2
Quad processor @ 2.66 GHz machine and MATLAB 7.9. In
the training phase, 10 frontal face images, with different
expressions, of each of the 100 subjects are selected as
training data set. For each face class (subject), two subclasses
are formed; one subclass contains the face images with
varying illumination, while the other subclass contains the
face images of the same subject with varying facial
expressions. In the testing phase, randomly chosen 200 face
images of the Texas 3D face database with variations in facial
expressions are used. The sample training images which are
used for our experimentation are shown in the Figure 2, and
their corresponding texture images are shown in the Figure 3.
The performance comparison of the proposed method with
the RT+PCA, RT+PCA+LDA and RT+Symbolic PCA[18],
in terms of recognition accuracy is presented in the Table 1.
The graph of recognition rates versus the number of
eigenfaces is shown in the Figure 4 for the proposed method
(RT+Symbolic LDA). It is observed that the recognition rate
improves as the number of eigenfaces is increased. It is
99.60% for 5 LDA components using SVM classifier in case
of the proposed method. Further, the proposed method based
on RT and Symbolic LDA outperforms the PCA, RT+PCA,
RT+PCA+LDA and RT+Symbolic PCA [18] methods. The
Figure 5 shows the receiver operating characteristic (ROC)
curve for the proposed method, where x-axis and y-axis
denote FAR and FRR, for Bhosphorus, Texas and CASIA 3D
face databases with equal error rate 13.3436, 11.3529 and
10.0113 respectively.

5. CONCLUSION

In this paper, we have proposed a novel method for three
dimensional (3D) face recognition using Radon transform
and Symbolic LDA based features of 3D range face images. In
this method, the Symbolic LDA based feature computation
takes into account face image variations to a larger extent and
has advantage of dimensionality reduction. The experimental
results have yielded 99.60% recognition accuracy using SVM
classifier with reduced complexity and a small number of
features, which compares well with other state-of-the-art
methods. The experimental results demonstrate the efficacy
and the robustness of the method to illumination and pose
variations. The recognition accuracy can be further improved
by considering a larger training set and a better classifier.

ACKNOWLEDGEMENT

The authors are grateful to the referees for their helpful
comments and suggestions. Also, the authors are indebted
to the University Grants Commission, New Delhi, for the
financial support for this research work under UGC-MRP
F.No.39-124/2010 (SR)
P. S. Hiremath et al., International Journal of Advances in Computer Science and Technology, 2(12), December 2013, 267-274

272




Figure 1. Overview of the proposed framework



Figure 2. Sample range images of the training set.



Figure 3. The facial texture images corresponding to the training range images of the Figure 2


Table 1. Performance comparison of the proposed with the RT+PCA, RT+PCA+LDA and
RT+Symbolic PCA, in terms of recognition accuracy.

No. of
Eigenfaces
Recognition Accuracy (in % ) Average
Time
Taken
For
Recognition
(in Secs.)
RT+PCA
[18]
RT +
Symbolic
PCA[21]
RT +
PCA +
LDA[18]
RT+Symbolic LDA
(Proposed Method)
Minimum
Distance
Classifier
KNN
(K=5)
SVM
5 60.10 62.00 61.60 63.46 63.96 64.46 9.265
10 77.50 78.00 77.90 80.06 80.46 81.46 9.292
15 84.36 85.50 85.10 87.46 87.76 87.96 9.405
20 90.19 91.00 91.20 93.46 93.96 93.96 9.421
25 94.10 95.00 94.20 96.56 96.96 97.16 9.727
30 94.16 96.00 95.91 97.46 98.46 98.96 10.166
35 95.20 96.50 97.90 98.46 98.46 99.26 10.216
40 95.30 97.00 99.16 99.50 99.56 99.60 11.105

P. S. Hiremath et al., International Journal of Advances in Computer Science and Technology, 2(12), December 2013, 267-274

273



Figure 4. The graph showing Performance comparison of the proposed with the RT+PCA, RT+PCA+LDA and
RT+Symbolic PCA, in terms of recognition accuracy



Figure 5. Receiver operating characteristic (ROC) curve for the proposed method,
where x-axis and y-axis denote FAR and FRR, for Bhosphorus, Texas and CASIA 3D face databases
with equal error rate 13.3436, 11.3529 and 10.0113 respectively.



REFERENCES

[1] R. Chellappa, C. Wilson, and S. Sirohey, Human and
machine recognition of faces: A survey, Proc. Of the IEEE,
vol. 83, no. 5, pp.704-740(May 1995).
[2] W. Zhao, R. Chellappa, P. J . Phillips, A. Rosenfeld, Face
Recognition: A Literature Survey, ACM Computing Surveys,
Vol. 35, No. 4, pp. 399458(December 2003).
[3] Patil A.M., Kolhe S.R. and Patil P.M., 2D Face Recognition
Techniques: A Survey, International J ournal of Machine
Intelligence, ISSN: 09752927, Volume 2, Issue 1,
pp.74-83(2010).
[4] Andrea F. Abate, Michele Nappi, Daniel Riccio, Gabriele
Sabatino, 2D and 3D face recognition: A survey, Pattern
Recognition Letters 28, pp.18851906, (2007).
[5] Kevin W. Bowyer, Kyong Chang, Patrick Flynn, A survey of
approaches and challenges in 3D and multi-modal 3D +2D
face recognition, Computer Vision and Image Understanding
101, 115 (2006).
[6] Shalini Gupta, Mia K. Markey, Alan C. Bovik,
Anthropometric 3D Face Recognition, Int. J . Computer Vis.,
Springer Science+Business Media, LLC (2010).
[7] Xiaoguang Lu , Dirk Colbry , Anil K. J ain, Matching 2.5D
Scans for Face Recognition, Int. Conf. Pattern Recog. (ICPR
2004), pp.362-366.
P. S. Hiremath et al., International Journal of Advances in Computer Science and Technology, 2(12), December 2013, 267-274

274


[8] Kyong I. Chang , Bowyer, K.W., Flynn P. J ., Adaptive Rigid
Multi-region Selection for Handling Expression Variation in
3D Face Recognition, Computer Vision and Pattern
Recognition - Workshops, 2005.
[9] J ahanbim, S., Hyohoon Choi , J ahanbin R., Bovik A. C.,
Automated facial feature detection and face recognition
using Gabor features on range and portrait images, 15th
IEEE International Conference on Image Processing (ICIP
2008).
[10] N. Alyz, B. Gkberk, H. Dibekliolu, A. Savran, A. A. Salah,
L. Akarun, B. Sankur, "3D Face Recognition Benchmarks on
the Bosphorus Database with Focus on Facial Expressions" ,
The First COST 2101 Workshop on Biometrics and Identity
Management (BIOID 2008), Roskilde University, Denmark,
May 2008.
[11] S. Gupta, K. R. Castleman, M. K. Markey, A. C. Bovik,
Texas 3D Face Recognition Database, IEEE Southwest
Symposiumon Image Analysis and Interpretation, May 2010,
p 97-100, Austin, TX. URL:
http://live.ece.utexas.edu/research/texas3dfr/index.htm.
[12] ChenghuaXu, Yunhong Wang, Tieniu Tan and Long Quan,
Automatic 3D Face Recognition Combining Global
Geometric Features with Local Shape Variation Information,
Proc. The 6th IEEE International Conference on Automatic
Face and Gesture Recognition (FG), pp.308-313,
2004.(CASIA 3D Face Database)
[13] Amir Averbuch and Yoel Shkolnisky, 3D Fourier based
discrete Radon transform, Appl. Comput. Harmon. Anal. 15 ,
Elsevier Inc. (2003), pp. 3369.
[14] M. Turk, A. Pentland, Eigenfaces for Recognition, J ournal
of Cognitive Neurosicence, Vol. 3, No. 1, 1991, pp. 71-86
[15] H. Moon, P.J . Phillips, Computational and Performance
aspects of PCA-based Face Recognition Algorithms,
Perception, Vol. 30, pp. 303-321(2001).
[16] K. Etemad, R. Chellappa, Discriminant Analysis for
Recognition of Human Face Images, J ournal of the Optical
Society of America A, Vol. 14, No. 8, August 1997, pp.
1724-1733.
[17] W. Zhao, R. Chellappa, A. Krishnaswamy, Discriminant
Analysis of Principal Components for Face Recognition, Proc.
of the 3rd IEEE International Conference on Face and Gesture
Recognition, FG'98, 14-16 April 1998, Nara, J apan, pp.
336-341.
[18] P. S. Hiremath and Manjunath Hiremath, Linear
discriminant analysis for 3D face recognition using radon
transform, International Conference on Multimedia,
Processing, Communication and Computing Applications,
PESIT, Bangalore, Springer LNEE, 13-15 December 2012.
[19] Hengliand Tang, Yanfeng Sun, Baocai Yin and Yun Ge, 3D
Face recognition based on Sparse representation, J ournal of
Supercomputing, Vol. 58, Issue 1, pp.84-95(2011).
[20] Bock, H. H. Diday E. (Eds) : Analysis of Symbolic Data,
Springer Verlag (2000).
[21] P. S. Hiremath and Manjunath Hiremath, 3D Face
Recognition using Radon Transformand Symbolic PCA,
International J ournal of Electronics and Computer Science
Engineering (IJ ECSE), ISSN 2277-1956, Vol.1,
No.4,pp.2342-2349
.


Dr. P. S. Hiremath was born in May 1952, and has obtained
Ph.D. (1978) in Applied Mathematics and M.Sc. (1973) in
Applied Mathematics, from Karnataka
University, Dharwad, Karnataka, India. He had
been in the Faculty of Mathematics and
Computer Science of various institutions in
India, namely, National Institute of
Technology, Surathkal (1977-79), Coimbatore
Institute of Technology, Coimbatore (1979-80), National
Institute of Technology, Tiruchirapalli (1980-86), Karnataka
University, Dharwad (1986-1993) and has been presently
working as Professor of Computer Science in Gulbarga
University, Gulbarga (1993 onwards). His research areas of
interest are Computational Fluid Dynamics, Optimization
Techniques, Image Processing and Pattern Recognition, and
Computer Networks. He has published 156 research papers in
peer reviewed International J ournals and Proceedings of
International Conferences.


Manjunatha Hiremath was born in J uly 1984, and has
obtained M.Phil (2010) in Computer Science and M.Sc.
(2008) in Computer Science fromGulbarga
University, Gulbarga. Presently, he is working
as Project Fellow in UGC Major Research
Project since February 2011. His area of
research interest is Image Processing and
Pattern Recognition. He has published 8
research papers in peer reviewed International J ournals and
Proceedings of International Conferences.



.